lecture1-introduction to medical image computingbagci/teaching/mic17/lec1.pdf• imaging sciences is...
Post on 22-May-2020
3 Views
Preview:
TRANSCRIPT
MEDICAL IMAGE COMPUTING (CAP 5937)- SPRING 2017
LECTURE 1: Introduction
Dr. Ulas BagciHEC 221, Center for Research in Computer Vision (CRCV), University of Central Florida (UCF), Orlando, FL 32814.bagci@ucf.edu or bagci@crcv.ucf.edu
1
• This is a special topics course, offered for the second time in UCF.
2
CAP5937: Medical Image Computing
Lorem Ipsum Dolor Sit Amet
• This is a special topics course, offered for the second time in UCF.
• Lectures: Mon/Wed, 10.30am-11.45am
3
CAP5937: Medical Image Computing
Lorem Ipsum Dolor Sit Amet
• This is a special topics course, offered for the second time in UCF.
• Lectures: Mon/Wed, 10.30am-11.45am
• Office hours: Mon/Wed, 1pm-2.30pm
4
CAP5937: Medical Image Computing
Lorem Ipsum Dolor Sit Amet
• This is a special topics course, offered for the second time in UCF.
• Lectures: Mon/Wed, 10.30am-11.45am
• Office hours: Mon/Wed, 1pm-2.30pm
• No textbook is required, materials will be provided.
• Avg. grade was A- last spring.
5
CAP5937: Medical Image Computing
Lorem Ipsum Dolor Sit Amet
6
Medical Image
Computing
Image Processing Computer
Vision
Machine Learning
Imaging Sciences
(Radiology, Biomedical)
Motivation• Imaging sciences is experiencing a tremendous
growth in the U.S. The NYT recently ranked biomedical jobs as the number one fastest growing career field in the nation and listed bio-medical imaging as the primary reason for the growth.
7
Motivation• Imaging sciences is experiencing a tremendous
growth in the U.S. The NYT recently ranked biomedical jobs as the number one fastest growing career field in the nation and listed bio-medical imaging as the primary reason for the growth.
• Biomedical imaging and its analysis are fundamental to (1) understanding, (2) visualizing, and (3) quantifying information.
8
Motivation• Imaging sciences is experiencing a tremendous
growth in the U.S. The NYT recently ranked biomedical jobs as the number one fastest growing career field in the nation and listed bio-medical imaging as the primary reason for the growth.
• Biomedical imaging and its analysis are fundamental to (1) understanding, (2) visualizing, and (3) quantifying information.
• This course will mostly focus on analysis of biomedical images, and imaging part will be briefly taught!
9
Syllabus• Basics of Radiological Image Modalities and their
clinical use (MRI, PET, CT, fMRI, DTI, …)• Introduction to Medical Image Computing and Toolkits• Image Filtering, Enhancement, Noise Reduction, and
Signal Processing• Medical Image Registration• Medical Image Segmentation• Medical Image Visualization• Machine Learning in Medical Imaging• Shape Modeling/Analysis of Medical Images
10
Syllabus• Grading:
– In-class Quiz (20%, approximately 20 quizzes, at the end of each lecture)
– 3 Programming Assignments (each 10%, total 30%)• ITK/VTK packages should be used• ITK and VTK provide necessary codes/libraries for medical image
processing and analysis. • C/C++ or Python can be used and call ITK/VTK functions• In-class collaboration is encouraged, but individual submission is
required.– 1 Individual Project (50%)
• Will be selected from a list of projects or you can come withh your own project
• A short presentation (15%), coding/method (25%), results (10%)
11
Optional Reading List• Image Processing, Analysis, and Machine Vision. M. Sonka, V.
Hlavac, R. Boyle. Nelson Engineering, 2014.• Level-set Methods, by J. A. Sethian, Cambridge University Press.• Visual Computing for Medicine: Theory, Algorithms, and
Applications. B. Preim, C. Botha. Morgan Kaufmann, 2013.• Medical Image Registration. J. Hajnal, D. Hill, D. Hawkes (eds).
CRC Press, 2001.• Pattern Recognition and Machine Learning. C. Bishop. Springer,
2007.• Insight into Images: Principles and Practice for Segmentation,
Registration and Image Analysis, Terry S. Yoo (Editor) (FREE)• Algorithms for Image Processing and Computer Vision, J. R. Parker• Medical Imaging Signals and Systems, by Jerry Prince & Jonathan
Links, Publisher: Prentice Hall
12
Conferences and Journals to Follow• The top-tier conferences (double blind, acceptance rates are below
25%, high quality technical articles):– MICCAI (medical image computing & computer assisted intervention)– IPMI (Information Processing in Medical Imaging)– Other conferences: IEEE ISBI, EMBC and SPIE Med Imaging– Clinical Conferences: RSNA (>65.000 attendances), ISMRM, SNM
• The top-tier technical journals:– IEEE TMI, TBME, PAMI, and TIP– Medical Image Analysis, CMIG, and NeuroImage
• The top-tier clinical journals relevant to MIC:– Radiology, Journal of Nuclear Medicine, AJR, Nature Methods, Nature
Medicine, PlosOne, …
13
Required skill set• Basic programming experience (any language is fine)• Linear Algebra/Matrix Algebra• Differential Equations• Basic Statistics
14
Biomedical Images• (Bio)medical images are different from other
pictures
15
Biomedical Images• (Bio)medical images are different from other
pictures– They depict distributions of various physical features
measured from the human body (or animal body).
16
Biomedical Images• (Bio)medical images are different from other
pictures– They depict distributions of various physical features
measured from the human body (or animal body).• Analysis of biomedical images is guided by very
specific expectations
17
Biomedical Images• (Bio)medical images are different from other pictures
– They depict distributions of various physical features measured from the human body (or animal).
• Analysis of biomedical images is guided by very specific expectations– Automatic detection of tumors, characterizing their types,– Measurement of normal/abnormal structures,– Visualization of anatomy, surgery guidance, therapy
planning, – Exploring relationship between clinical, genomic, and
imaging based markers
18
Free Software to use in this course• ImageJ (and/or FIJI)• ITK-Snap• SimpleITK• MITK• FreeSurfer• SLICER• OsiriX• An extensive list of software: www.idoimaging.com and
blue: will be frequently used in this course
19
Medical Image Formats• Dicom• Nifti• Analyze (img/hdr)• Raw data• …
20
DICOM (the mostly used)• Digital Imaging and Communications in Medicine standard• Since its first publication in 1993, DICOM has revolutionized
the practice of radiology, allowing the replacement of X-ray film with a fully digital workflow.
• It is the international standard for medical images and related information (ISO 12052)
• defines the formats for medical images that can be exchanged with the data and quality necessary for clinical use.
• It is implemented in almost every radiology, cardiology imaging, and radiotherapy device (X-ray, CT, MRI, ultrasound, etc.), and increasingly in devices in other medical domains such as ophthalmology and dentistry.
21
3D Slicer Software• A software platform for the analysis (including
registration and interactive segmentation) andvisualization (including volume rendering) ofmedical images and for research in image guidedtherapy.
• A free, open source software available on multipleoperating systems: Linux, MacOSX and Windows
• Extensible, with powerful plug-in capabilities foradding algorithms and applications.
22
Brief History of 3D Slicer• 1997: Slicer started
as a research project between the Surgical Planning Lab (Harvard) and the CSAIL (MIT)
• Open Source + Open Data + Open Community
23
•80 authorized developers contributing to the source code of Slicer
Slicer Volume Module
24
Slicer Welcome Module
25
Slicer Welcome Module
26
Loading A DICOM Volume
27
Loading A DICOM Volume
28
Loading A DICOM Volume
29
Loading A DICOM Volume
30
Interactive exploration
31
32
Interactive exploration
33
3D Slicer Sources• http://slicer.org/• http://www.slicer.org/slicerWiki/index.php/Docum
entation/UserTraining• https://vimeo.com/37671358
34
Libraries to be Used• ITK and VTK
– National Library of Medicine Insight Segmentation and Registration Toolkit (ITK).
– ITK is an open-source, cross-platform system that provides developers with an extensive suite of software tools for image analysis.
– C/C++, Python, Matlab, …
35
Goals of ITK– Supporting the Visible Human Project. – Establishing a foundation for future research. – Creating a repository of fundamental algorithms. – Developing a platform for advanced product
development. – Support commercial application of the technology. – Create conventions for future work. – Grow a self-sustaining community of software users and
developers.
36
History of ITK• ITK was initially conceived by the NLM(National
Library of Medicine).• An initiative for open source software tools to
analyze human dataset• Developed by group of both commercial and
academic organizations(kitware, GE research, Mathsoft, Upenn, UT, UNC
• Goal: provide a foundation to enable research in image processing and biomedical image computing, Providing catalog of algorithms
37
What is ITK?• Image Processing
• Segmentation
• Registration
• No Graphical User Interface (GUI)
• No Visualization
38
Coordinate System for Reading Files• Multiple coordinate frames
– Physical– Patient– Index
• ITK uses LPS (Left Posterior Superior) for DICOM
39
How to Integrate ITK in application
Credit: itk.org
Installation/Requirements
C++ Compiler
gcc 2.95 – 3.3Visual C++ 6.0Visual C++ 7.0Visual C++ 7.1
Intel 7.1Intel 8.0IRIX CC
Borland 5.5Mac – gcc
CMakewww.cmake.org
Credit: itk.org
Installation process• Google ITK , go to the download page, download
the zip file (or directly install using github or console functions in mac/linux)
• Google cmake, go to the download page, get the binaries and install the binaries.
Configuring ITK – MS-Windows
l Run CMake
l Select the SOURCE directory
l Select the BINARY directory
l Select your Compiler
Configuring ITK
Configuring ITKl Disable BUILD_EXAMPLES
l Disable BUILD_SHARED_LIBS
l Disable BUILD_TESTING
l Click “Configure” to configure
l Click “OK” to generate project files
Building ITKl Open ITK.sln in the Binary Directory
l Select ALL_BUILD project
l Build it
…It will take about 15 minutes …
Verify the BuiltLibraries will be found in
ITK_BINARY / bin / { Debug, Release }• The following libraries should be there:
– ITKCommon
– ITKBasicFilters
– ITKAlgorithms
– ITKNumerics
– ITKFEM
ITKIO
ITKStatistics
ITKMetaIO
itkpng
itkzlib
Use ITK from an external Project
Copy “HelloWorld.cxx”“CMakeLists.txt”
from the Examples/Installation
Directoryinto another
directory
RunCMake
• Select Source Dir• Select Binary Dir
Use ITK from an external Project
• accept the default in
CMAKE_BACKBARD_COMPATIBILITY
• leave empty EXECUTABLE_OUTPUT_PATH
• leave empty LIBRARY_OUTPUT_PATH
• Set ITK_DIR to the binary directory
where ITK was built
Build Sample Project• Open HelloWorld.sln generated by CMake
• Select ALL_BUILD project
• Build it
Run the example• Locate the file HelloWorld.exe
• Run it…
• It should produce the message:
ITK Hello World !
Starting your own project• Create a clean new directory
• Write a CMakeLists.txt file
• Write a simple .cxx file
• Configure with CMake
• Build
• Run
Writing CMakeLists.txtPROJECT( myProject )
FIND_PACKAGE ( ITK )IF ( ITK_FOUND )
INCLUDE( ${ITK_USE_FILE} )ENDIF( ITK_FOUND )
ADD_EXECUTABLE( myProject myProject.cxx )
TARGET_LINK_LIBRARIES ( myProject ITKCommon ITKIO)
Writing myProject.cxx
#include "itkImage.h"#include "itkImageFileReader.h"#include "itkGradientMagnitudeImageFilter.h"
int main( int argc, char **argv ) {typedef itk::Image<unsigned short,2> ImageType;typedef itk::ImageFileReader<ImageType> ReaderType;typedef itk::GradientMagnitudeImageFilter<
ImageType,ImageType> FilterType;
ReaderType::Pointer reader = ReaderType::New();FilterType::Pointer filter = FilterType::New();
reader->SetFileName( argv[1] );filter->SetInput( reader->GetOutput() );filter->Update();return 0;
}
Run CMake
How to find what you need?
• Follow the link Alphabetical List
• Follow the link Groups
• Post to the insight-users mailing list
http://www.itk.org/ItkSoftwareGuide.pdf
http://www.itk.org/Doxygen/html/index.html
VTK• The Visualization Toolkit (VTK) is an open-
source, freely available software system for 3D computer graphics, image processing, and visualization.
• Consists of a C++ class library and several interpreted interface layers including Tcl/Tk, Java, and Python
VTK• Download VTK from vtk.org• Configure VTK
– Run CMake– Select the SOURCE directory– Select the BINARY directory– Select your Compiler (same used for ITK)
Configuring VTK
Disable• BUILD_EXAMPLES• BUILD_SHARED
Leave unchanged• CMAKE_BACKWARD_COMPATIBILITY• VTK_DATA_ROOT
Enable• VTK_USE_HYBRID• VTK_USE_RENDERING• VTK_USE_PARALLEL• VTK_USE_PATENTED
Disable• VTK_WRAP_JAVA• VTK_WRAP_PYTHON• VTK_WRAP_TCLEnable (Advanced)• VTK_USE_ANSI_STDLIB
Build VTK• Open VTK.dswin the Binary Directory• Select ALL_BUILD project• Build it …It may take about 90 minutes …
Verify the BuildLibraries will be found in
VTK_BINARY / bin/ { Debug, Release}
Verify the BuildThe following libraries
should be there:• vtkCommon• vtkFiltering• vtkImaging• vtkGraphics• vtkHybrid• vtkParallel• vtkPatented
• Vtkexpat• Vtkfreetype• Vtkftgl• Vtkjpeg• Vtkpng• Vtktiff• vtkzlib
Starting your own projectwith ITK + VTK
• Create a clean new directory• Write a CmakeLists.txtfile• Write a simple .cxxfile• Configure with CMake• Build• Run
Writing CMakeLists.txt
PROJECT(myProject)
FIND_PACKAGE ( ITK)IF ( ITK_FOUND)INCLUDE( ${USE_ITK_FILE} )ENDIF( ITK_FOUND)
FIND_PACKAGE ( VTK)IF ( VTK_FOUND)INCLUDE( ${USE_VTK_FILE} )ENDIF( VTK_FOUND)
(continue...)
Writing CMakeLists.txt
INCLUDE_DIRECTORIES(${myProject_SOURCE_DIR})
ADD_EXECUTABLE( myProject myProject.cxx)
TARGET_LINK_LIBRARIES ( myProjectITKBasicFiltersITKCommonITKIOvtkRenderingvtkGraphicsvtkHybridvtkImagingvtkIOvtkFilteringvtkCommon)
Writing myProject.cxx
Writing myProject.cxx#include "itkImage.h"#include "itkImageFileReader.h"#include "itkImageToVTKImageFilter.h"#include "vtkImageViewer.h"#include "vtkRenderWindowInteractor.h"intmain( intargc, char **argv) {typedef itk::Image<unsigned short,2> ImageType;typedef itk::ImageFileReader<ImageType> ReaderType;typedef itk::ImageToVTKImageFilter<ImageType> connectorType;ReaderType::Pointer reader= ReaderType::New();ConnectorType::Pointer connector= ConnectorType::New();
Writing myProject.cxx
reader->SetFileName( argv[1]);connector->SetInput( reader->GetOutput() );vtkImageViewer* viewer= vtkImageViewer::New();vtkRenderWindowInteractor* renderWindowInteractor= vtkRenderWindowInteractor::New();viewer->SetupInteractor( renderWindowInteractor);viewer->SetInput( connector->GetOutput() );viewer->Render();viewer->SetColorWindow( 255);viewer->SetColorLevel( 128);renderWindowInteractor->Start();return 0;}
SimpleITK• New Wrapper for the insight segmentation &
registration toolkit• Goal :to help rapid prototyping and expand the user
based of ITK by exposing the algorithms to new users
• Simplify the algorithms so they don’t depend on types of images.
• Binary built in distributions• Supports 2D & 3D image , multi component images• Easy importing & exporting • data through Numpy
• QUESTIONS?
77
Quiz!(1 pt)
78
top related